基于改进Census变换的局部立体匹配算法  

Improved Census Transform for Local Stereo Match

作　　者：赵杰[1,2] 张肖帅 赵茏菲 郭莉莉[1,2] 贺光美 

机构地区：[1]河北大学电子信息工程学院,河北保定071000 [2]河北省数字医疗工程重点实验室,河北保定071000

出　　处：《电视技术》2015年第11期6-10,共5页Video Engineering

基　　金：河北省卫生厅科研基金项目(20120395);河北省教育厅科学技术研究重点项目(ZD20131086);河北大学中西部高校提升综合实力工程项目

摘　　要：针对基于传统Census变换的立体匹配算法鲁棒性差和精度不高的问题,提出一种基于改进Census变换的自适应权重立体匹配算法。首先,用Census变换窗口中邻域像素的中值来替换中心像素的灰度值,克服了邻域像素对中心像素的依赖;然后用自适应权重的方法分别计算匹配代价和进行立体匹配,得到初始视差;最后通过左右一致性校验和亚像素提精的方法得到稠密的视差图。实验结果表明,该算法有很强的鲁棒性和很高的匹配精度。In order to improve the robustness and accuracy of stereo matching based on Census transform,a adaptive weighting stereo matching algorithm based on improved Census transform is proposed.Firstly,the gray value of central pixel is replaced by the median of neighborhood pixels of the transform window,which overcomes the dependence of neighborhood pixels on the center pixel;secondly,the initial parallax is obtained by using the adaptive weighting algorithm on the cost calculation and stereo matching;lastly,the last disparity map is obtained by means of left-right cheking and sub-pixel enhancement.The results demonstrate that this method is more robust and accurate.

关 键 词：立体匹配 Census变换 自适应权重 中值 左右一致性校验 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]